Forecast infrastructure cost elasticity

What it's for

Answers the executive question behind cloud and infrastructure budgets: when demand doubles, does cost scale sublinearly, linearly, or accelerate past a capacity breakpoint—and did that model work on later periods?

What you give it

Inputs split into evidence read from your connected systems, calibration your team owns, and numerical controls that affect precision but never the result's meaning.

{
  "future_scenarios": [
    {
      "id": "base",
      "price_index_by_period": [
        1.4,
        1.4,
        1.4
      ],
      "probability": 0.7,
      "workload_units_by_period": [
        40,
        60,
        100
      ]
    },
    {
      "id": "growth",
      "price_index_by_period": [
        1.4,
        1.4,
        1.4
      ],
      "probability": 0.3,
      "workload_units_by_period": [
        80,
        120,
        160
      ]
    }
  ],
  "historical_periods": [
    {
      "id": "infrastructure-period-00",
      "observed_cost": 44.58686746157203,
      "period": 1,
      "price_index": 1,
      "workload_units": 20
    },
    {
      "id": "infrastructure-period-01",
      "observed_cost": 57.43585499831026,
      "period": 2,
      "price_index": 1.01,

What you get back

This is the actual output of running the example above — computed by the same function the platform calls, not an illustration.

{
  "assumptions": [
    "Historical periods are chronological point-in-time observations with comparable workload, fully loaded infrastructure cost and a positive price index; the final holdout is untouched by breakpoint selection and coefficient fitting.",
    "A continuous one-breakpoint log-log response adequately represents real cost elasticity over the supported workload range. Extrapolation, architecture regime changes, vendor tariff discontinuities, commitments, capacity ceilings and correlated residuals require separate scenario treatment.",
    "The piecewise candidate is selected on an internal future block, refit before the untouched holdout, and compared with a constant-unit-cost baseline. Failure of workload variation, holdout improvement or interval coverage is an abstention even when a numerical forecast is returned.",
    "Elasticity is predictive association, not causal proof that demand created cost, a universal benchmark, a vendor-efficiency score or authority to resize, migrate, terminate, purchase or change staffing."
  ],
  "candidate_model_diagnostics": [
    {
      "breakpoint_workload_units": null,
      "selection_mean_absolute_log_error": 0.1177
    },
    {
      "breakpoint_workload_units": 30,
      "selection_mean_absolute_log_error": 0.0706
    },
    {
      "breakpoint_workload_units": 50,
      "selection_mean_absolute_log_error": 0
    },
    {
      "breakpoint_workload_units": 80,
      "selection_mean_absolute_log_error": 0.0728
    }
  ],
  "configuration": {
    "holdout_periods": 5,
    "internal_selection_periods": 7,
    "maximum_interval_coverage_error": 0.25,
    "minimum_holdout_relative_improvement": 0.1,
    "prediction_interval": 0.8,
    "price_index_normalization": "fit real cost as observed_cost divided by price_index; restore future nominal price index",
    "residual_log_standard_deviation": 0,
    "ridge_penalty": 0
  },
  "decision": "infrastructure_cost_elasticity_forecast_clears_validation_gates",
  "future_scenario_forecasts": [
    {
      "forecast_total_expected_cost": 1363.9904,
      "maximum_period_expected_cost": 695.5668,
      "probability": 0.3,
      "scenario_id": "growth",
      "sum_of_period_cost_p10": 1363.9886,
      "sum_of_period_cost_p50": 1363.9904,

How it works

Forecasting & survival — Estimate when something completes or fails, with censoring and unresolved work handled honestly rather than dropped.

1. 1 Build a chronological period ledger of positive workload, fully loaded infrastructure cost and price index, reserve the final holdout, and reserve an earlier future block inside training solely for response-shape selection.
2. 2 Compare constant-elasticity and train-derived breakpoint candidates on the internal selection block, ridge-fit the chosen continuous log-log response on all pre-holdout history, and estimate its residual interval without touching future scenarios.
3. 3 Require supported workload variation, material holdout error improvement over a constant-unit-cost baseline, and acceptable holdout interval coverage before using the elasticity and price-restored scenario cost forecast.

Before you trust it

Every tool in the catalog ships with the conditions under which its answer is meaningful — and the conditions under which it should abstain instead of guessing.

Calibrating it to your org

Calibration workflow

1. 1 Define the management decision, target outcome, aggregate unit, privacy boundary, cadence, and prediction/intervention horizon for this organization.
2. 2 Build a tenant-scoped historical cohort using only information available before each prediction or decision; preserve zero periods, censoring, assignment probabilities, and unresolved outcomes when the method requires them.
3. 3 Estimate statistical parameters on training history, but obtain costs, utilities, risk tolerance, practical-effect thresholds, capacity, and policy constraints from accountable decision owners.
4. 4 Validate on later time windows or held-out aggregate units at the deployment grain, against a simple baseline and the function-specific validation strategy.
5. 5 Deploy only if the returned decision clears evidence, overlap, calibration, robustness, and guardrail checks; warning, unsupported, schema-gap, and fallback decisions are abstentions.
6. 6 Monitor realized outcomes, data drift, coverage, and decision regret; recalibrate at a governed cadence or after a detected regime/definition change, never merely because a stakeholder dislikes the result.

Call it from your AI

You don't wire up 388 tools in your MCP client. The GitRevio MCP server exposes 18 tools, three of which let an agent search the catalog, read a tool's schema, and run it — so the assistant finds this one on its own.

gitrevio_capabilities_search
  { "q": "select a continuous piecewise loglog infrastructure" }
  → finds "forecast_infrastructure_cost_elasticity"

gitrevio_capability_describe
  { "capability_id": "forecast_infrastructure_cost_elasticity" }
  → returns the input schema and agent guidance shown on this page

gitrevio_capability_run
  { "capability_id": "forecast_infrastructure_cost_elasticity", "arguments": { ... } }
  → returns the result shown above

Works in Claude Desktop, Claude Code, Cursor, Cline, Continue.dev, Goose and Aider. See the MCP server.

Related tools

See every tool in Finance & unit economics →